Zipper:: a zero-infrastructure peer-to-peer system forvanet
Proceedings of the 3rd ACM workshop on Wireless multimedia networking and performance modeling
ALCA: agent learning---based clustering algorithm in vehicular ad hoc networks
Personal and Ubiquitous Computing
Hi-index | 0.00 |
It was recently noticed that co-directional roadways traffic consists of a collection of disconnected clusters, implying that end-to-end connectivity between sender and receiver is not guaranteed to exist. Because of this lack of connectivity, a car that stores a packet may have to carry it for a while, acting as a "data mule", before a suitable next hop can be identified. We offer an analytical expression for the expected size of a cluster in co-directional traffic, confirming the findings of [2, 3, 4]. Next, we develop TAPR -- a Traffic-Adaptive Packet Relaying protocol wherecars use a simple strategy to decide whether to route through co-directional or oncoming clusters. TAPR is evaluated through extensive simulation.